We examine the effects of U.S. federal land programs on private conservation using county-level panel regressions. Private conservation data measure acres held by The Nature Conservancy (TNC) and by participants in Land Trust Alliance (LTA) censuses. Government data measure federal landholdings (e.g., national parks and forests) and enrollment in the Conservation and Wetland Reserves (CRP and WRP). We find a small crowding-out effect from the CRP on LTA trusts. With TNC, we find crowding in from the CRP and crowding out from federal landholdings. Our theory gives insights as to why these and other effects (e.g., population and income) differ between TNC and the LTA trusts.